【摘要】 [目的/意义]开展科技领域技术演进多路径识别及其创新特性分析,可为技术起源探究和技术价值评估提供有效帮助。[方法/过程]构建科技领域专利引证网络,通过Louvain社团检测和SPC主路径提取进行技术演进多路径识别。在此基础上,利用技术—功效共词矩阵来探究各技术演进路径的核心技术演化,利用结构洞约束度指标和技术邻近度指标来识别各技术路径节点专利的创新类型。[结果/结论]利用德温特专利索引数据库中1963—2018年柔性显示领域的9121条专利记录进行实证研究,结果表明提出的方法能够较为全面地识别科技领域技术演进的多个路径方向及其核心技术,通过结构洞约束度和技术邻近度指标能够有效检测技术演进过程中的渐进性创新和突破性创新技术。更多还原

【Abstract】 [Purpose/significance] Identifying multi-path and innovation characteristics in the process of domain technology evolution can provide effective help for exploring the origin of technology and evaluating particular patents.[Method/process] Firstly, a patent citation network of the technology domain is constructed and the sub network communities are divided by community detection algorithm of Louvain.Then, the main paths of the whole citation network and each sub network communities are extracted respectively based on SPC algorithm to demonstrate multiple technology evolution paths.Finally, the technology efficacy co-word matrix is used to explore the evolution of core technologies, the indexes of structural-hole constraint and technology proximity is utilized to identify the innovation type of each patent in technology evolution paths.[Result/conclusion] An empirical study was conducted through 9121 patent records in the field of flexible display from 1963 to 2018 in Derwent patent index database.The results show that the proposed method can recognize multiple evolution paths of technology field, and that the incremental innovation and breakthrough innovation technology can be effectively detected by combing indexes of structural-hole constraint and technology proximity.更多还原